Graph-structured data appears frequently in domains including chemistry, natural language semantics, social networks, and knowledge bases. In this work, we study feature learning techniques for graph-structured inputs. Our starting point is previous work on Graph Neural Networks (Scarselli et al., 2009), which we modify to use gated recurrent units and modern optimization techniques and then extend to output sequences. The result is a flexible and broadly useful class of neural network models that has favorable inductive biases relative to purely sequence-based models (e.g., LSTMs) when the problem is graph-structured. We demonstrate the capabilities on some simple AI (bAbI) and graph algorithm learning tasks. We then show it achieves state-of-the-art performance on a problem from program verification, in which subgraphs need to be described as abstract data structures.

We have developed a one-step method to synthesize carbon quantum dots (CQD) from biogenic polyamines (PAs) as an antibacterial agent for topical treatment of bacterial keratitis (BK). CQDs synthesized by direct pyrolysis of spermidine (Spd) powder through a simple dry heating treatment exhibit a solubility and yield much higher than those from putrescine and spermine. We demonstrate that CQDs obtained from Spds (CQD) possess effective antibacterial activities against non-multidrug-resistant Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Salmonella enterica serovar Enteritidis bacteria and also against the multidrug-resistant bacteria, methicillin-resistant S. aureus. The minimal inhibitory concentration (MIC) of CQD is ∼2500-fold lower than that of spermidine alone, demonstrating their strong antibacterial capabilities. Investigation of the possible mechanisms behind the antibacterial activities of the as-synthesized CQD indicates that the super-cationic CQD with small size (diameter ca. 6 nm) and highly positive charge (ζ-potential ca. +45 mV) cause severe disruption of the bacterial membrane. In vitro cytotoxicity, hemolysis, hemagglutination, genotoxicity, and oxidative stress and in vivo morphologic and physiologic cornea change evaluations show the good biocompatibility of CQD. Furthermore, topical ocular administration of CQD can induce the opening of the tight junction of corneal epithelial cells, thereby leading to great antibacterial treatment of S. aureus-induced BK in rabbits. Our results suggest that CQD are a promising antibacterial candidate for clinical applications in treating eye-related bacterial infections and even persistent bacteria-induced infections.

The phytohormone salicylic acid (SA) plays essential roles in biotic and abiotic responses, plant development, and leaf senescence. 2,5-Dihydroxybenzoic acid (2,5-DHBA or gentisic acid) is one of the most commonly occurring aromatic acids in green plants and is assumed to be generated from SA, but the enzymes involved in its production remain obscure. DMR6 (Downy Mildew Resistant6; At5g24530) has been proven essential in plant immunity of Arabidopsis (Arabidopsis thaliana), but its biochemical properties are not well understood. Here, we report the discovery and functional characterization of DMR6 as a salicylic acid 5-hydroxylase (S5H) that catalyzes the formation of 2,5-DHBA by hydroxylating SA at the C5 position of its phenyl ring in Arabidopsis. S5H/DMR6 specifically converts SA to 2,5-DHBA in vitro and displays higher catalytic efficiency (K cat /K m = 4. ) for SA. Interestingly, S5H/DMR6 displays a substrate inhibition property that may enable automatic control of its enzyme activities. The s5h mutant and s5hs3h double mutant overaccumulate SA and display phenotypes such as a smaller growth size, early senescence, and a loss of susceptibility to Pseudomonas syringae pv tomato DC3000. S5H/DMR6 is sensitively induced by SA/pathogen treatment and is expressed widely from young seedlings to senescing plants, whereas S3H is more specifically expressed at the mature and senescing stages. Collectively, our results disclose the identity of the enzyme required for 2,5-DHBA formation and reveal a mechanism by which plants fine-tune SA homeostasis by mediating SA 5-hydroxylation.

No abstract

With the rapid development of data‐driven human interaction, advanced data‐storage technologies with lower power consumption, larger storage capacity, faster switching speed, and higher integration density have become the goals of future memory electronics. Nevertheless, the physical limitations of conventional Si‐based binary storage systems lag far behind the ultrahigh‐density requirements of post‐Moore information storage. In this regard, the pursuit of alternatives and/or supplements to the existing storage technology has come to the forefront. Recently, organic‐based resistive memory materials have emerged as promising candidates for next‐generation information storage applications, which provide new possibilities of realizing high‐performance organic electronics. Herein, the memory device structure, switching types, mechanisms, and recent advances in organic resistive memory materials are reviewed. In particular, their potential of fulfilling multilevel storage is summarized. Besides, the present challenges and future prospects confronted by organic resistive memory materials and devices are discussed.

Language modelling provides a step towards intelligent communication systems by harnessing large repositories of written human knowledge to better predict and understand the world. In this paper, we present an analysis of Transformer-based language model performance across a wide range of model scales -from models with tens of millions of parameters up to a 280 billion parameter model called Gopher. These models are evaluated on 152 diverse tasks, achieving state-of-the-art performance across the majority. Gains from scale are largest in areas such as reading comprehension, fact-checking, and the identification of toxic language, but logical and mathematical reasoning see less benefit. We provide a holistic analysis of the training dataset and model's behaviour, covering the intersection of model scale with bias and toxicity. Finally we discuss the application of language models to AI safety and the mitigation of downstream harms.

Leakage interference between memory cells is the primary obstacle for enlarging X‐point memory arrays. Metal‐filament threshold switches, possessing excellent selectivity and low leakage current, are developed in series with memory cells to reduce sneak path current and lower power consumption. However, these selectors typically have limited on‐state currents (≤10 µA), which are insufficient for memory RESET operations. Here, a strategy is proposed to achieve sufficiently large RESET current (≈2.3 mA) by introducing highly ordered Ag nanodots to the threshold switch. Compared to the Ag thin film case, Ag nanodots as active electrode could avoid excessive Ag atoms migration into solid electrolyte during operations, which causes stable conductive filament growth. Furthermore, Ag nanodots with rapid thermal processing contribute to forming multiple weak Ag filaments at a lower voltage and then spontaneous rupture as the applied voltage reduced, according to quantized conductance and simulation analysis. Impressively, the Ag nanodots based threshold switch, which is bidirectional and truly electroforming‐free, demonstrates extremely high selectivity >10 9 , ultralow leakage current <1 pA, very steep slope of 0.65 mV dec −1 , and good thermal stability up to 200 °C, and further represents significant suppression of leakage currents and excellent performances for SET/RESET operations in the one‐selector‐one‐resistor configuration.

This study reports a two-step method to synthesize spermidine-capped fluorescent carbon quantum dots (Spd-CQDs) and their potential application as an antibacterial agent. Fluorescent carbon quantum dots (CQDs) are synthesized by pyrolysis of ammonium citrate in the solid state and then modified with spermidine by a simple heating treatment without a coupling agent. Spermidine, a naturally occurring polyamine, binds with DNA, lipids, and proteins involved in many important processes within organisms such as DNA stability, and cell growth, proliferation, and death. The antimicrobial activity of the as-synthesized Spd-CQDs (size ≈4.6 nm) has been tested against non-multidrug-resistant E. coli, S. aureus, B. subtilis, and P. aeruginosa bacteria and also multidrug-resistant bacteria, methicillin-resistant S. aureus (MRSA). The minimal inhibitory concentration value of Spd-CQDs is much lower (>25 000-fold) than that of spermidine, indicating their promising antibacterial characteristics. The mechanism of antibacterial activity is investigated, and the results indicate that Spd-CQDs cause significant damage to the bacterial membrane. In vitro cytotoxicity and hemolysis analyses reveal the high biocompatibility of Spd-CQDs. To demonstrate its practical application, in vitro MRSA-infected wound healing studies in rats have been conducted, which show faster healing, better epithelialization, and formation of collagen fibers when Spd-CQDs are used as a dressing material.